Small Businesses Buy Islands. Enterprises Pay by the Bridge.

Ask what businesses actually want their software to touch, and the market splits into two species.

Three out of five small-business asks touch no other system at all. No sync, no connector, no data flowing anywhere else. They want the thing itself: the booking page, the invoice maker, the store. An island.

Enterprise demand is built the other way. Fewer than half of enterprise asks stand alone, and the ones that connect are not bigger by a little.

The ladder nobody puts on a pricing page

Take a standalone enterprise ask as the baseline. The same kind of ask touching one or two other systems is budgeted around three times higher. Three to five systems, around five times. Six or more, roughly eleven times the standalone version.

Each system an ask touches roughly doubles what the enterprise expects to pay for it.

And the connected asks are not just larger. They are stickier. At six or more systems, three in four enterprise requests want an ongoing relationship, not a delivery. The bridge buyer is the retention buyer.

What gets bridged is remarkably consistent: Stripe, where the money is. Google Analytics, where the proof is, the scoreboard we wrote about. Shopify, where the orders are. HubSpot, where the customers are. Enterprises are not asking for integrations in general. They are asking for the four or five systems their business actually lives in, stitched together.

The small-business inversion

Here is the part that trips builders: for small businesses, the ladder barely exists. A many-system ask pays only modestly more than an island, and, strangely, it is less likely to be ongoing. Around half of simple small-business asks want a standing arrangement. For the connected ones, fewer than a third.

Read that carefully. When a small business finally buys a bridge, they buy it as a one-off: wire my store to my accounting, then leave me alone. Connection is a chore to complete, not a capability to subscribe to.

So the same feature list means opposite things at the two ends of the market. Down-market, an integration is plumbing the buyer hopes to never see again. Up-market, the integration list is the product, and the buyer prices it connector by connector.

One product, two architectures

• Ship the island down-market. First-session usefulness, no setup, nothing to connect before value appears. The small buyer is not withholding budget until you integrate. They are withholding attention until you work alone.
• When small businesses do connect, make it set-and-forget. One-time wiring to the two systems they actually name, the money and the scoreboard: Stripe and Google Analytics. "Connected out of the box" beats "integration platform" at this end, every time.
• Sell the bridge up-market, and price it like the data does. If each connected system doubles what an enterprise expects to pay, your integration tier is not a feature checkbox. It is the pricing page. Most builders charge for seats while their enterprise buyer is silently pricing connections.
• Watch the tide. Multi-system asks from small businesses have grown by about a quarter recently, while the enterprise share has eased. The island buyers are starting to ask for bridges, small ones, prewired. Whoever packages the two-connector island first meets that demand as it arrives.

The market is not asking whether your product integrates. It is asking two different questions in two different accents: does it work alone, and what does it connect. Answering the right question to the right buyer is worth more than most feature roadmaps.

This reading comes from millions of demand signals, collected continuously. Not surveys and not commentary, but real buying activity, the same intelligence that runs inside the product. One more instance of what we keep finding since the market started moving at the speed of AI: the buyers did not get more alike. They got easier to tell apart.